Background: This study aims to evaluate the diagnostic performance of Deep Learning (DL) machine for the detection of adenomyosis on uterine ultrasonographic images and compare it to intermediate ultrasound skilled trainees. Methods: Prospective observational study were conducted between 1 and 30 April 2022. Transvaginal ultrasound (TVUS) diagnosis of adenomyosis was investigated by an experienced sonographer on 100 fertile-age patients. Videoclips of the uterine corpus were recorded and sequential ultrasound images were extracted. Intermediate ultrasound-skilled trainees and DL machine were asked to make a diagnosis reviewing uterine images. We evaluated and compared the accuracy, sensitivity, positive predictive value, F1-score, specificity and negative predictive value of the DL model and the trainees for adenomyosis diagnosis. Results: Accuracy of DL and intermediate ultrasound-skilled trainees for the diagnosis of adenomyosis were 0.51 (95% CI, 0.48–0.54) and 0.70 (95% CI, 0.60–0.79), respectively. Sensitivity, specificity and F1-score of DL were 0.43 (95% CI, 0.38–0.48), 0.82 (95% CI, 0.79–0.85) and 0.46 (0.42–0.50), respectively, whereas intermediate ultrasound-skilled trainees had sensitivity of 0.72 (95% CI, 0.52–0.86), specificity of 0.69 (95% CI, 0.58–0.79) and F1-score of 0.55 (95% CI, 0.43–0.66). Conclusions: In this preliminary study DL model showed a lower accuracy but a higher specificity in diagnosing adenomyosis on ultrasonographic images compared to intermediate-skilled trainees.
Background: to compare several uterine biometric parameters at transvaginal ultrasound (TVUS) between adenomyosis and non-adenomyosis uteri and evaluate their role for the diagnosis of diffuse adenomyosis. Methods: prospective observational study conducted between the 1 February 2022 and the 30 April 2022. In this case, 56 patients with TVUS diagnosis of adenomyosis were included. A 1:1 ratio age and parity-matched group of non-adenomyosis patients was selected. We compared sonographic uterine biometric parameters (longitudinal (LD), anteroposterior (APD) and transverse (TD) diameters, volume, simple and complex diameter ratios) and investigated their diagnostic performance. Results: all sonographic parameters were significantly different between the study groups, except for TD/(LD+APD). Optimal cut-off values of APD and LD/APD showed the best sensitivity and specificity. APD diameter equal or superior to 39.5 mm (95% CI, 36.2–42.8) had sensitivity of 0.70 (95% CI, 0.57–0.80), specificity of 0.71 (95% CI, 0.59–0.82) and accuracy of 0.75 (95% CI, 0.66–0.84). LD/APD equal or inferior to 2.05 (95% CI, 1.96–2.13) showed sensitivity and specificity of 0.70 (95% CI, 0.57–0.80) each and accuracy of 0.72 (95% CI, 0.62–0.81). Conclusions: several biometric uterine parameters at TVUS in fertile-aged women were statistically different between adenomyosis and non-adenomyosis uteri, though their optimal cut-off values showed low accuracy in diagnosing adenomyosis.
Background: We evaluated the efficacy of local methotrexate (MTX) treatment followed by hysteroscopic resection for caesarean scar pregnancy and its impact on future fertility. Methods: Monocentric, prospective, observational study performed in the Haykel Hospital between June 2016 and December 2020. Twenty-one women with caesarean scar pregnancy underwent a transcutaneous ultrasound-guided direct injection of MTX into the gestational sac in an outpatient setting. Hysteroscopic resection of residual trophoblastic retention was then performed according to perisaccular blood flow. Main results: Two patients had complete spontaneous trophoblast expulsion after MTX injection, and hysteroscopy was performed in 19 patients for residual trophoblastic retention 1 to 12 weeks after MTX injection. Successful preservation of a healthy uterus with the combined procedure was obtained in 94.8% of patients. Hemostatic hysterectomy was required in one patient. Mean hospitalization duration was 1.5 days. Three patients had spontaneous pregnancy after the procedure. Conclusion: Direct MTX injection into the gestational sac for caesarean scar pregnancy followed by hysteroscopic resection was an effective technique with a short hospitalization, fertility preservation and a low major complication rate compared with other modalities of treatment reported in the literature. Further larger prospective comparative studies are needed to confirm the efficacy of this procedure.
Endometriosis is a common benign gynecological disease characterized by the presence of endometrial glands and stroma outside the uterus. It can be defined as endometrioma, superficial peritoneal endometriosis or deep infiltrating endometriosis (DIE) depending on the location and the depth of infiltration of the organs. In 5%–12% of cases, DIE affects the digestive tract, frequently involving the distal part of the sigmoid colon and rectum. Surgery is generally recommended in cases of obstructive symptoms and in cases with pain that is non‐responsive to medical treatment. Selection of the most optimal surgical technique for the treatment of bowel endometriosis must consider different variables, including the number of lesions, eventual multifocal lesions, as well as length, width and grade of infiltration into the bowel wall. Except for some major and widely accepted indications regarding bowel resection, established international guidelines are not clear on when to employ a more conservative approach like rectal shaving or discoid resection, and when, instead, to opt for bowel resection. Damage to the pelvic autonomic nervous system may be avoided by detection of the middle rectal artery, where its relationship with female pelvic nerve fibers allows its use as an anatomical landmark. To reduce the risk of potential vascular and nervous complications related to bowel resection, a less invasive approach such as shaving or discoid resection can be considered as potential treatment options. Additionally, the middle rectal artery can be used as a reference point in cases of upper bowel resection, where a trans mesorectal technique should be preferred to prevent devascularization and denervation of the bowel segments not affected by the disease.
Purpose: To evaluate the diagnostic performance of Deep Learning (DL) machine for the detection of adenomyosis on uterine ultrasonographic images and compare it to intermediate ultrasound skilled trainees. Methods: Prospective observational study conducted between 1st and 30th April 2022. Transvaginal ultrasound (TVUS) diagnosis of adenomyosis was investigated by an experienced sonographer on 100 fertile-age patients. Videoclips of the uterine corpus were recorded and sequential ultrasound images were extracted. Intermediate ultrasound skilled trainees and DL machine were asked to make a diagnosis reviewing uterine images. We evaluated and compared the accuracy, sensitivity, positive predictive value, F1- score, specificity and negative predictive value of the DL model and the trainees for adenomyosis diagnosis. Results: Accuracy of DL and intermediate ultrasound skilled trainees for the diagnosis of adenomyosis were 0.51 (95% CI, 0.48-0.54) and 0.70 (95% CI, 0.60-0.79), respectively. Sensitivity, specificity and F1-score of DL were 0.43 (95% CI, 0.38-0.48), 0.82 (95% CI, 0.79-0.85) and 0.46 (0.42-0.50), whereas intermediate ultrasound skilled trainees had sensitivity of 0.72 (95% CI, 0.52-0.86), specificity of 0.69 (95% CI, 0.58-0.79) and F1-score of 0.55 (95% CI, 0.43-0.66). Conclusion: In this preliminary study DL model showed a lower accuracy but a higher specificity in diagnosing adenomyosis on ultrasonographic images compared to intermediate skilled trainees.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.